Graph Neural Network contextual embedding for Deep Learning on Tabular Data
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2023
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866911785479569408 |
|---|---|
| author | Villaizán-Vallelado, Mario Salvatori, Matteo Martinez, Belén Carro Esguevillas, Antonio Javier Sanchez |
| author_facet | Villaizán-Vallelado, Mario Salvatori, Matteo Martinez, Belén Carro Esguevillas, Antonio Javier Sanchez |
| contents | All industries are trying to leverage Artificial Intelligence (AI) based on their existing big data which is available in so called tabular form, where each record is composed of a number of heterogeneous continuous and categorical columns also known as features. Deep Learning (DL) has constituted a major breakthrough for AI in fields related to human skills like natural language processing, but its applicability to tabular data has been more challenging. More classical Machine Learning (ML) models like tree-based ensemble ones usually perform better. This paper presents a novel DL model using Graph Neural Network (GNN) more specifically Interaction Network (IN), for contextual embedding and modelling interactions among tabular features. Its results outperform those of a recently published survey with DL benchmark based on five public datasets, also achieving competitive results when compared to boosted-tree solutions. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2303_06455 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Graph Neural Network contextual embedding for Deep Learning on Tabular Data Villaizán-Vallelado, Mario Salvatori, Matteo Martinez, Belén Carro Esguevillas, Antonio Javier Sanchez Machine Learning Artificial Intelligence All industries are trying to leverage Artificial Intelligence (AI) based on their existing big data which is available in so called tabular form, where each record is composed of a number of heterogeneous continuous and categorical columns also known as features. Deep Learning (DL) has constituted a major breakthrough for AI in fields related to human skills like natural language processing, but its applicability to tabular data has been more challenging. More classical Machine Learning (ML) models like tree-based ensemble ones usually perform better. This paper presents a novel DL model using Graph Neural Network (GNN) more specifically Interaction Network (IN), for contextual embedding and modelling interactions among tabular features. Its results outperform those of a recently published survey with DL benchmark based on five public datasets, also achieving competitive results when compared to boosted-tree solutions. |
| title | Graph Neural Network contextual embedding for Deep Learning on Tabular Data |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2303.06455 |